Bridging the Gap between Labeled and Unlabeled Data via Unified Flow with Feature Memory Bank
2026-08-17 • Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionArtificial Intelligence
AI summaryⓘ
The authors address a problem in semi-supervised semantic segmentation where labeled data tends to overshadow unlabeled data, leading to poor quality pseudo-labels. They propose a method called UFFM that combines a visual foundation model with a remote sensing teacher model to better train on both labeled and unlabeled data together. Their approach also uses a memory bank to keep track of features per class, helping to align labeled and unlabeled data features. Experiments on remote sensing datasets show their method improves over existing techniques by better balancing and optimizing data usage.
semi-supervised learningsemantic segmentationpseudo-labelsremote sensingvisual foundation modelfeature memory bankclass-feature alignmentunified trainingdomain adaptationfeature representation
Authors
Shanwen Wang, Xin Sun, Danfeng Hong, Junyu Dong, Patrick Le Callet
Abstract
Although semi-supervised semantic segmentation ($\text{S}^4$) utilizes abundant unlabeled data to reduce manual labeling burdens, independent training of labeled and unlabeled data causes the former to dominate, which severely degrades pseudo-label quality. To address this challenges, we propose a novel remote sensing (RS) $\text{S}^4$ method via unified flow with feature memory bank (UFFM). Specifically, UFFM comprises two key innovations: unified flow (UF) and feature memory bank (FMB). The UF is a new training flow that generates less biased pseudo-labels by combining an external visual foundation model (VFM) with an RS domain teacher, and jointly optimizes labeled and pseudo-labeled data under a unified training objective. The FMB is a novel memory module for $\text{S}^4$ that dynamically updates class-specific features during training and reduces the feature discrepancy between labeled and unlabeled data through class-feature alignment. To verify the effectiveness of our model, we conduct extensive experiments on RS datasets. The experimental results show the superiority of our method over SOTA $\text{S}^4$ methods. Moreover, the results demonstrate the effectiveness of our contributions in bridging the optimization and feature representation gap between labeled and unlabeled data. Our code is released at \href{https://github.com/wangshanwen001/RS-UFFM}{https://github.com/wangshanwen001/RS-UFFM}.